Can I create an I or D controller in LabVIEW?

[aeroAggie]

I successfully tuned a PID Advanced VI to contol a process. Through trial and error, I found that the following gains worked best for me: Kc = 0.5, Ti = 0.01, Td = 0. Now, I would like to fiddle with a D controller; however, I don't think I can do that since the PID Advanced VI uses the standard form to calculate gains.

 Do I have to create my own "ideal" PID VI, or is there a way to create an I or D controller using the PID and PID Advanced VIs?

Many thanks!

-Farid

[Sam_Sharp]

I think there is a VI somewhere (control design?) that allows you to convert the gains between ideal and standard form?

[CoastalMaineBird]

I don't understand your question - the VI you're using is already a D controller (as well as P and I).

D is usually the trickiest to get right (for me, anyway).

Can you restate your trouble?

[aeroAggie]

The PID Advanced VI uses the "noninteractive" algorithm seen in the picture below. This algorithm does not have individual Kp, Ki, and Kd gains; rather, the algorithm uses a controller gain, Kc, and integral and derivative periods, Ti and Td respectively, to control a process (which, according to the pic's site, is more intuitive and commonly used in industry).

http://blog.opticontrols.com/archives/124

However, this means that I can only create a P, PI, PD, or PID controller. I want to fiddle with a D controller. Can I do that with the PID or PID Advanced VIs? If so, how?

Thanks,
Farid

[aeroAggie]

Found it! PID Structure Conversion VI

But a lone Ki or Kd term still won't affect my PID Advanced vi's output (with Ki and Kd ranging from 5E-7 to 5E7). Am I wiring the PID Advanced VI incorrectly?

-Farid

[William1225]

I have the same question that I'd like to do something like I-control, but it seemed the p-value is necessary in PID toolkit...

As a result, I made the p-value as small as possible and got the corresponding I-value I need.

I'm not sure my methond is correct, and looking for some other ideas.

My objective is to control a temperature chamber by PID control.

For example, the chamber is set 37 degC, but the sensor we're interested in is 36 degC.

I use 37 degC as the SV, and the measured value of the sensor as the PV, and I have to generate an output to set the chamber dynamically.

What my idea is I can get the accumulated integral error as an offset to the SV so that I add the output of the I-control with the SV as a actual output to set the chamber. 

P value is trivial to me in this way since when the error is zero, the output is zero, and have no offset to the output, like the following image.